Unverified50% confidenceFactExact time
PanoPed-S contains 108,000 frames of synthetic data from three perspectives: fixed cameras, quadruped robot-mounted, and drone-mounted.
1
Sources
50%
Confidence
Long-term
Relevance
10/8/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedEvery frame of PanoPed-S is annotated with synchronized segmentation masks, depth information, camera poses, and 3D pedestrian state labels.78% similarUnverifiedThe research team constructed a dedicated dataset, SPW-NavSet, containing panoramic videos with camera trajectories and verified instructions.67% similarUnverifiedA Reddit user publicly shared a dataset captured with a highly specular-reflective "mirror suit" robot costume, comprising 425 images in RAW/JPEG format65% similarUnverifiedThe project is a pose-based teleoperation system that captures human body pose in real time via an ordinary webcam and maps it onto the joints of a quadruped robot.65% similarUnverifiedVisDrone is a large-scale drone vision dataset released by the AISKYEYE team at Tianjin University, containing more than 260,000 frames of images and over 10,000 video clips, with annotations for 10 categories including pedestrians, vehicles, and bicycles.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/986879API
curl https://kongchang.com/api/v1/knowledge/claims/986879MCP
get_claim(id=986879)